Flight delay risk early warning and coping system based on big data analysis

Abstract

The invention belongs to the technical field of aviation management, and discloses a flight delay risk early warning and coping system based on big data analysis. Comprising a data acquisition module used for acquiring aviation operation data; the data analysis module is used for acquiring route flow; the delay prediction module is used for predicting the flight delay probability; the delay early warning module is used for judging whether an early warning instruction is generated or not according to the flight delay probability, and predicting flight delay duration if the early warning instruction is generated; the claim payment module is used for calculating the claim payment amount for claim payment based on the flight delay duration, and judging whether to perform meal payment or not; the travel and accommodation arrangement module is used for judging whether arrangement information is generated or not based on the flight delay duration; according to the method, big data and artificial intelligence technologies are fully utilized, intelligent prediction and refined service of flight delay risks are realized, the operation efficiency and service efficiency of airlines are improved, and the travel experience and satisfaction of passengers are greatly improved.

Description

Technical Field

[0001] The present invention relates to the field of aviation management technology, and more specifically, to a flight delay risk warning and response system based on big data analysis. Background Art

[0002] With the rapid development of the global aviation industry and the continuous growth of passenger travel demand, flight delays have become a major challenge affecting airlines' operational efficiency and passengers' travel experience. According to statistics from the International Air Transport Association (IATA), flight delays not only cause additional operating costs for airlines, but also significantly reduce passenger satisfaction and trust, and affect the reputation of airline brands. Traditional manual management and single emergency response methods can no longer meet the needs of modern aviation operations. Especially in the case of large-scale flight delays, how to quickly and accurately provide passengers with compensation, meals, accommodation and transportation arrangements has become a key factor in improving service quality and operational management efficiency.

[0003] With the maturity of technologies such as big data, artificial intelligence, the Internet of Things and cloud computing, the aviation industry is exploring how to deal with flight delays through intelligent and digital means; the patent with publication number CN107657394A discloses a flight delay processing system and method, including: a server-side device and a mobile-side device, wherein: the server-side device is used to publish flight delay information to the mobile-side device; the mobile-side device is used to receive the flight delay information and generate meal arrangement information, car arrangement information and hotel arrangement information according to the flight delay information; this invention solves the existing technical problems of easy errors and low efficiency caused by manual recording, and achieves the technical effect of simply and efficiently handling flight delays;

[0004] However, although the above technology involves relevant service arrangements during flight delays, it only briefly mentions concepts such as generating meal arrangement information, car arrangement information and hotel arrangement information, without elaborating on the specific generation steps and detailed content of this information; this general description makes it difficult for passengers to obtain timely and comprehensive service support during flight delays in actual applications, and their personalized needs cannot be accurately met, which directly affects the passengers' travel experience and satisfaction; at the same time, the lack of a clear service generation process and information management system will also lead to inaccurate or untimely service execution, reducing customer satisfaction during flight delays, and further affecting the evaluation of the airline's brand image and its service quality.

[0005] In view of this, the present invention proposes a flight delay risk warning and response system based on big data analysis to solve the above problems. Summary of the invention

[0006] In order to overcome the above-mentioned defects of the prior art and to achieve the above-mentioned purpose, the present invention provides the following technical solution: a flight delay risk warning and response system based on big data analysis, comprising:

[0007] Data collection module, used to collect aviation operation data;

[0008] Data analysis module, used to analyze aviation operation data and obtain route traffic;

[0009] Delay prediction module, which is used to analyze aviation operation data and route traffic to predict the probability of flight delays;

[0010] The delay warning module determines whether to generate a warning instruction based on the probability of flight delay. If a warning instruction is generated, the flight delay duration is predicted;

[0011] The compensation payment module calculates the compensation amount based on the flight delay duration and determines whether to pay meals.

[0012] The accommodation arrangement module determines whether to generate arrangement information based on the flight delay duration.

[0013] Further, the aviation operation data includes flight data and weather data;

[0014] Flight data collection methods include:

[0015] Collect target flight data, which is the flight information of the target flight; the target flight is the flight for which flight delay prediction is performed, and the flight information includes take-off time, landing time, flight route and flight distance, the take-off time is the time when the flight is scheduled to take off from the departure airport, the landing time is the time when the flight is scheduled to land at the arrival airport, the flight route is the flight path between the departure airport and the arrival airport of the flight, and the flight distance is the flight distance corresponding to the flight route; obtain the current time, and mark the take-off time in the target flight data as the target take-off time; according to the current time and the target take-off time, obtain related flights, which are all flights whose take-off times are between the current time and the target take-off time; according to the related flights, obtain related flight data, which is the flight information of the related flights;

[0016] Mark the flight routes in the target flight data as target routes, and mark the flight routes in the related flight data as related routes; obtain a world map, draw the target route and the related routes on the world map, and determine whether there is a route in the related routes that intersects with the target route; mark the routes in the related routes that intersect with the target route as related routes, and do not mark the routes in the related routes that do not intersect with the target route; mark the flight corresponding to the related route as a related flight, mark the flight information of the related flight as related flight data, and use the related flight data and the target flight data as flight data;

[0017] The weather data is the weather information corresponding to the departure airport at the target take-off time, including wind speed, precipitation, visibility and temperature.

[0018] Furthermore, the method for obtaining route traffic includes:

[0019] Mark the take-off time in the associated flight data as the associated take-off time, and mark the landing time as the associated landing time; convert the current time, target take-off time, associated take-off time, and associated landing time into a unified unit; subtract the corresponding associated take-off time from the associated landing time of each associated flight to obtain the estimated flight time of each associated flight; subtract the associated take-off time of each associated flight from the target take-off time to obtain the elapsed flight time of each associated flight; divide the elapsed flight time of each associated flight by the corresponding estimated flight time, and then multiply by the corresponding flight distance to obtain the elapsed flight distance of each associated flight;

[0020] Get the departure airport and landing airport of each associated flight. According to the world map, get the longitude and latitude corresponding to the departure airport and landing airport of each associated flight. The longitude and latitude include longitude and latitude. Mark the longitude and latitude corresponding to the departure airport as the starting longitude and latitude, and mark the longitude and latitude corresponding to the landing airport as the ending longitude and latitude. Subtract the corresponding starting longitude from the ending longitude of each associated flight to obtain the longitude difference. Subtract the corresponding starting latitude from the ending latitude of each associated flight to obtain the latitude difference. Calculate the current longitude and latitude corresponding to each associated flight according to the longitude difference, latitude difference, flown distance and flight distance corresponding to each associated flight. The expression of the current longitude is: ; In the formula, is the current longitude, is the starting point longitude, is the distance flown, is the flight distance, is the longitude difference; the expression for the current latitude is: ; In the formula, is the current latitude, is the starting point latitude, is the latitude difference;

[0021] Divide the world map into airspace according to the preset division scheme to obtain airspace, is an integer greater than 1; according to the current longitude and latitude of each associated flight, mark the position of each associated flight on the world map; mark the airspace passed by the target route on the world map as the target airspace, count the number of associated flights in each target airspace, and mark it as airspace traffic; take the airspace traffic corresponding to all target airspaces as route traffic.

[0022] Furthermore, the method for predicting the probability of flight delay includes:

[0023] Use weather data and route traffic as analysis data, input the analysis data into the trained delay prediction model, and predict the corresponding flight delay probability;

[0024] The method for determining whether to generate an early warning instruction includes:

[0025] Set a warning threshold and compare the flight delay probability with the warning threshold; if the flight delay probability is greater than or equal to the warning threshold, generate a warning instruction; if the flight delay probability is less than the warning threshold, do not generate a warning instruction;

[0026] The step of predicting flight delay duration includes:

[0027] Step 1: Obtain historical traffic, which is the route traffic obtained at historical time; take the airspace traffic corresponding to the same target airspace in the historical traffic as a set of traffic sets, and the traffic set corresponds to the target airspace one by one;

[0028] Step 2: Input each traffic set into the trained traffic prediction model to predict the airspace traffic at the future time T, where T is an integer greater than 1;

[0029] Step 3: Take all predicted airspace traffic as predicted traffic, obtain weather data at future times, predict the corresponding flight delay probability based on the weather data at future times and predicted traffic, and determine whether to generate an early warning instruction;

[0030] Step 4: If a warning instruction is still generated, replace the earliest airspace traffic in each group of traffic sets with the corresponding predicted airspace traffic, and return to step 2; if no warning instruction is generated, go to step 5;

[0031] Step 5: Get the number of warning instructions generated in step 3 and mark it as the delay number. Multiply the delay number by T to get the flight delay duration. The unit of the flight delay duration is consistent with the unit of the target take-off time.

[0032] Furthermore, in step 2, the traffic prediction model is an RNN neural network model, and the training process of the traffic prediction model includes:

[0033] Pre-collect the same airspace continuously airspace traffic and build a traffic training set. , is the number of airspace flows in a set of flow sets; based on the flow training set, a flow prediction model for predicting airspace flows at future moments is trained;

[0034] Preset the sliding step size L and the sliding window length H; use the sliding window method to convert the airspace traffic in the traffic training set into multiple training samples, each training sample includes H airspace traffic; use each training sample as the input of the traffic prediction model, predict the airspace traffic after the sliding step size L as the output, and use the actual L airspace traffic corresponding to each training sample as the prediction target. Use the mean absolute percentage error MAPE to evaluate the model accuracy of the prediction results. When the calculated MAPE is less than the preset MAPE, the traffic prediction model training corresponding to the airspace traffic is completed; generate a traffic prediction model that predicts the airspace traffic at future times based on the airspace traffic.

[0035] Furthermore, the method for calculating the compensation amount and making compensation includes:

[0036] Preset compensation standards, including compensation amounts corresponding to different flight delay durations; obtain corresponding compensation amounts according to the compensation standards and flight delay durations; obtain passenger payment information corresponding to the target flight, including payment type and payment account number, and pay the compensation amount to all passengers corresponding to the target flight according to the passenger payment information corresponding to the target flight;

[0037] The method for determining whether to distribute meals includes:

[0038] Preset the delivery threshold and compare the flight delay duration with the delivery threshold; if the flight delay duration is greater than or equal to the delivery threshold, the meal will be delivered; if the flight delay duration is less than the delivery threshold, the current time will be added to the flight delay duration to obtain the estimated departure time; preset meal time, meal time includes breakfast time , lunch time and dinner time ,in, ; Compare the estimated departure time with the meal time; if , meals will be distributed, including is the estimated time of departure, means or; if , no meals will be distributed;

[0039] If meals are distributed, the mobile phone number of each target passenger is obtained and marked as a push number. The target passengers are passengers corresponding to the target flights. Electronic meal coupons are pushed to each target passenger based on the push number of each target passenger.

[0040] Further, the method for determining whether to generate arrangement information includes:

[0041] A scheduling threshold is preset, and the estimated take-off time is compared with the scheduling threshold; if the estimated take-off time is greater than or equal to the scheduling threshold, scheduling information is generated; if the estimated take-off time is less than the scheduling threshold, scheduling information is not generated;

[0042] The arrangement information includes transportation arrangement information and accommodation arrangement information;

[0043] Methods for generating traffic arrangement information include:

[0044] Obtain the passenger information data of each target passenger, which includes age, cabin type, check-in counter and check-in time, and the unit of check-in time is consistent with the unit of current time; divide all target passengers into travel groups according to the check-in counter and check-in time of each target passenger, where if only one passenger travels, it is also considered as a travel group; subtract the check-in time of each target passenger from the current time to obtain the waiting time of each target passenger; calculate the priority corresponding to each travel group according to the age, cabin type and waiting time of each target passenger; divide all travel groups into group sets according to the priority of each travel group; set digital labels for each group set in increasing order and mark them as group labels; obtain the single round-trip time, which is the time it takes for the shuttle vehicle to pick up the passenger to the accommodation and then return to the terminal; multiply each single round-trip time by the group label corresponding to each group set, and add the current time to obtain the boarding time corresponding to each group set; convert the boarding time corresponding to each group set into a standard time format, and use it as traffic arrangement information, and push the traffic arrangement information corresponding to each group set to the corresponding target passenger.

[0045] Furthermore, the method for dividing all target passengers into travel groups includes:

[0046] The target passengers corresponding to each check-in counter are regarded as a group of check-in sets, and the check-in sets correspond to the check-in counters one by one; each target passenger in each check-in set is sorted from large to small according to the corresponding check-in time, and a check-in sorting table corresponding to each check-in set is generated; the difference between each two adjacent check-in times in each check-in sorting table is calculated in turn as the check-in time difference; a time difference threshold is preset, and each check-in time difference is compared with the time difference threshold respectively; if the check-in time difference is less than or equal to the time difference threshold, the two target passengers corresponding to the corresponding check-in time difference are regarded as a travel group; if the check-in time difference is greater than the time difference threshold, the two target passengers corresponding to the corresponding check-in time difference are regarded as a travel group respectively; all travel groups are analyzed, and if there are the same target passengers in two travel groups, the travel groups with the same target passengers are merged until there are no same target passengers in all travel groups;

[0047] The method for calculating the priority corresponding to each travel group includes:

[0048] Set different digital numbers for different cabin types and mark them as cabin numbers; preset age thresholds and , , ; Compare the age of each target passenger with the age threshold and For comparison; if , then the age coefficient assigned to the corresponding target passenger is 0, is the age of the target passenger; if , then the age coefficient assigned to the corresponding target passenger is ;like , then the age coefficient assigned to the corresponding target passenger is ;

[0049] The expression of the priority corresponding to the target passenger is: ; In the formula, For priority, is the age coefficient, is the cabin number, The waiting time is , , are all preset weight coefficients; the number of target passengers in each travel group is counted and marked as the number of passengers; the priorities of all target passengers in each travel group are added up in turn, and then divided by the corresponding number of passengers to obtain the priority corresponding to each travel group.

[0050] Furthermore, the method of dividing all travel groups into group sets includes:

[0051] Get the number of shuttle vehicles and the vehicle passenger capacity, where the vehicle passenger capacity is the maximum number of passengers that a single shuttle vehicle can transport; sort each travel group from large to small according to the corresponding priority to generate a group sorting table; replace each travel group in the group sorting table with the corresponding number of passengers; add each passenger number in the group sorting table to all the passengers in front of the corresponding passenger number to obtain the total number of passengers corresponding to each passenger number, and replace each passenger number in the group sorting table with the corresponding total number of passengers; multiply the number of shuttle vehicles by the vehicle passenger capacity to obtain the maximum single passenger capacity; mark the total number of passengers with the largest value in the group sorting table as the maximum number of passengers, divide the maximum number of passengers by the maximum single passenger capacity, and round up to obtain the number of connections; multiply the maximum single passenger capacity by the increasing connection coefficient to obtain The overall maximum passenger capacity; where the connection coefficient is an integer and the range of the connection coefficient is , n is the number of connections; The total maximum passenger capacity is added to the group sorting table, and the travel group corresponding to the total number of passengers between every two total maximum passenger capacities is taken as a group set to obtain n group sets; the range of group labels is .

[0052] Furthermore, the method for generating accommodation arrangement information includes:

[0053] Count the number of people in each travel group who are younger than the age threshold The target number of passengers is obtained and marked as the exclusion number; the number of passengers in each travel group is subtracted from the corresponding exclusion number to obtain the number of accommodations corresponding to each travel group; the number of accommodations for each travel group is divided by 2 to obtain the number of rooms corresponding to each travel group; each travel group in the group sorting table is replaced with the corresponding number of rooms, and the corresponding rooms are arranged for each travel group in positive order, and the room number corresponding to each travel group is obtained; the room number corresponding to each travel group is used as the accommodation arrangement information, and the accommodation arrangement information corresponding to each travel group is pushed to the corresponding target passengers.

[0054] Technical effects and advantages of the flight delay risk warning and response system based on big data analysis of the present invention:

[0055] By collecting and analyzing aviation operation data, we can accurately obtain route traffic and use deep learning technology to predict the probability of flight delays in real time, so as to issue timely flight delay risk warnings; according to the generated warning instructions, we can provide accurate compensation, meal distribution, transportation arrangements and accommodation arrangements to effectively respond to flight delays and solve the problem that traditional manual management and single emergency response methods cannot accurately meet passenger needs; make full use of big data and artificial intelligence technology to realize intelligent prediction of flight delay risks and refined services, which not only improves the operating efficiency and service efficiency of airlines, but also greatly improves passengers' travel experience and satisfaction, providing strong support for the intelligent management of the aviation industry. BRIEF DESCRIPTION OF THE DRAWINGS

[0056] Figure 1 This is a schematic diagram of a flight delay risk warning and response system based on big data analysis according to Embodiment 1 of the present invention;

[0057] Figure 2 This is a flow chart of a method for predicting flight delay duration according to Embodiment 1 of the present invention. DETAILED DESCRIPTION

[0058] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0059] Example 1

[0060] See also Figure 1 As shown, the flight delay risk warning and response system based on big data analysis described in this embodiment includes a data collection module, a data analysis module, a delay prediction module, a delay warning module, a compensation payment module and a travel and accommodation arrangement module; each module is connected by wired and / or wireless means to achieve data transmission between modules;

[0061] Data collection module, used to collect aviation operation data.

[0062] Aviation operations data includes flight data and weather data;

[0063] Flight data collection methods include:

[0064] Collect target flight data, which is the flight information of the target flight; the target flight is the flight for which flight delay prediction is performed, and the flight information includes take-off time, landing time, flight route and flight distance, the take-off time is the time when the flight is scheduled to take off from the departure airport, the landing time is the time when the flight is scheduled to land at the arrival airport, the flight route is the flight path between the departure airport and the arrival airport of the flight, and the flight distance is the flight distance corresponding to the flight route; the target flight data is obtained through the official website of the airline corresponding to the target flight; the current time is obtained, and the take-off time in the target flight data is marked as the target take-off time; according to the current time and the target take-off time, the relevant flights are obtained, and the relevant flights are all flights whose take-off time is between the current time and the target take-off time; the relevant flights are obtained by traversing the official websites of all airlines; according to the relevant flights, the relevant flight data is obtained, and the relevant flight data is the flight information of the relevant flights; the relevant flight data is obtained through the official website of the airline corresponding to the relevant flights;

[0065] Mark the flight routes in the target flight data as target routes, and mark the flight routes in the related flight data as related routes; obtain a world map through a map API, where the world map is a plan view showing countries, regions, and geographical features on the earth's surface; draw the target route and related routes on the world map, and determine whether there is a route in the related routes that intersects with the target route; mark the routes in the related routes that intersect with the target route as related routes, and do not mark the routes in the related routes that do not intersect with the target route; mark the flights corresponding to the related routes as related flights, mark the flight information of the related flights as related flight data, and use the related flight data and the target flight data as flight data;

[0066] Weather data refers to the weather information corresponding to the target take-off time at the departure airport, including wind speed, precipitation, visibility and temperature. Weather data is obtained from the official website of the Meteorological Bureau.

[0067] Data analysis module, used to analyze aviation operation data and obtain route traffic;

[0068] Methods for obtaining route traffic include:

[0069] The take-off time in the associated flight data is marked as the associated take-off time, and the landing time is marked as the associated landing time; the current time, the target take-off time, the associated take-off time, and the associated landing time are all converted into a unified unit. The preferred unit in this embodiment is minutes. For example, Convert to minutes; subtract the corresponding associated take-off time from the associated landing time of each associated flight to obtain the estimated flight time of each associated flight; subtract the associated take-off time of each associated flight from the target take-off time to obtain the elapsed flight time of each associated flight; divide the elapsed flight time of each associated flight by the corresponding estimated flight time, and then multiply by the corresponding flight distance to obtain the elapsed flight distance of each associated flight;

[0070] Get the departure airport and landing airport of each associated flight. According to the world map, get the longitude and latitude corresponding to the departure airport and landing airport of each associated flight. The longitude and latitude include longitude and latitude. The departure airport and landing airport of each associated flight can be obtained through the official website of the airline corresponding to the associated flight. Mark the longitude and latitude corresponding to the departure airport as the starting longitude and latitude, and mark the longitude and latitude corresponding to the landing airport as the ending longitude and latitude. Subtract the corresponding starting longitude from the ending longitude of each associated flight to obtain the longitude difference. Subtract the corresponding starting latitude from the ending latitude of each associated flight to obtain the latitude difference. Calculate the current longitude and latitude corresponding to each associated flight according to the longitude difference, latitude difference, flown distance and flight distance corresponding to each associated flight. The expression of the current longitude is: ; In the formula, is the current longitude, is the starting point longitude, is the distance flown, is the flight distance, is the longitude difference; the expression for the current latitude is: ; In the formula, is the current latitude, is the starting point latitude, is the latitude difference;

[0071] Divide the world map into airspace according to the preset division scheme to obtain airspace, is an integer greater than 1; the division scheme is preset by those skilled in the art according to actual conditions; the location of each associated flight is marked on the world map according to the current latitude and longitude of each associated flight; the airspace passed by the target route on the world map is marked as the target airspace, the number of associated flights in each target airspace is counted and marked as airspace flow; the airspace flow corresponding to all target airspaces is taken as route flow;

[0072] Delay prediction module, which is used to analyze aviation operation data and route traffic to predict the probability of flight delays;

[0073] Methods for predicting the probability of flight delays include:

[0074] Weather data and route traffic are used as analysis data, and the analysis data is input into the trained delay prediction model to predict the corresponding flight delay probability; the training process of the delay prediction model includes:

[0075] Pre-collection Group analysis data, Each group of analysis data is set with the corresponding flight delay probability. is an integer greater than 1, and the analysis data and the corresponding flight delay probability are converted into a corresponding set of feature vectors; the flight delay probability corresponding to the analysis data is collected by technical personnel in the historical flight delay risk warning process. Group analysis data, under the condition of the same group analysis data, judge whether the flight is delayed multiple times, divide the number of flight delays by the number of times the flight is judged to be delayed, and obtain the flight delay probability corresponding to the corresponding analysis data, and so on. Analyze the group analysis data to obtain the corresponding flight delay probability. The group analysis data sets the corresponding flight delay probabilities in turn;

[0076] Each set of feature vectors is used as the input of the delay prediction model. The delay prediction model uses a set of predicted flight delay probabilities corresponding to each set of analysis data as output, and the actual flight delay probability corresponding to each set of analysis data as the prediction target. The actual flight delay probability is the pre-set flight delay probability corresponding to the analysis data. Minimizing the sum of the prediction errors of all analysis data is used as the training target. The calculation formula of the prediction error is: ; In the formula, is the prediction error, is the group number of the eigenvector corresponding to the analyzed data, For the The predicted flight delay probability corresponding to the group analysis data, For the The actual flight delay probability corresponding to the group analysis data is trained, and the delay prediction model is trained until the sum of the prediction errors reaches convergence.

[0077] The above-mentioned delay prediction model is specifically a deep neural network model; it includes an input layer, a hidden layer and an output layer; each hidden layer includes multiple neurons, each neuron is connected to the neurons in the next layer, and the connection contains weights, which determine the importance and influence of data transmission in the neural network; an activation function is applied to each neuron between the hidden layer and the output layer, and the activation function introduces nonlinearity, allowing the network to learn more complex patterns and features.

[0078] The delay warning module determines whether to generate a warning instruction based on the probability of flight delay. If a warning instruction is generated, the flight delay duration is predicted;

[0079] Methods for determining whether to generate an early warning instruction include:

[0080] A warning threshold is set, and the warning threshold is pre-set by a technician in this field according to actual conditions; the flight delay probability is compared with the warning threshold; if the flight delay probability is greater than or equal to the warning threshold, a warning instruction is generated, indicating that the target flight will be delayed and flight delay service needs to be provided; if the flight delay probability is less than the warning threshold, no warning instruction is generated, indicating that the target flight will take off normally and will not be delayed;

[0081] like Figure 2 As shown in Figure 2, the steps for predicting flight delay duration include:

[0082] Step 1: Obtain historical traffic, which is the route traffic obtained at historical time; take the airspace traffic corresponding to the same target airspace in the historical traffic as a set of traffic sets, and the traffic set corresponds to the target airspace one by one;

[0083] Step 2: Input each traffic set into the trained traffic prediction model to predict the airspace traffic at the future time T, where T is an integer greater than 1; in this embodiment, the future time T is preferably the future T minutes;

[0084] Step 3: Take all predicted airspace traffic as predicted traffic, obtain weather data at future times, predict the corresponding flight delay probability based on the weather data at future times and predicted traffic, and determine whether to generate an early warning instruction;

[0085] Step 4: If a warning instruction is still generated, replace the earliest airspace traffic in each group of traffic sets with the corresponding predicted airspace traffic, and return to step 2; if no warning instruction is generated, go to step 5;

[0086] Step 5: Get the number of warning instructions generated in step 3 and mark it as the delay number. Multiply the delay number by T to get the flight delay duration. The unit of the flight delay duration is consistent with the unit of the target take-off time.

[0087] In the above step 2, the traffic prediction model is an RNN neural network model. The training process of the traffic prediction model includes:

[0088] Pre-collect the same airspace continuously airspace traffic and build a traffic training set. , is the number of airspace flows in a set of flow sets; based on the flow training set, a flow prediction model for predicting airspace flows at future moments is trained;

[0089] According to the actual experience of those skilled in the art, the sliding step size L and the sliding window length H are preset; the airspace traffic in the traffic training set is converted into multiple training samples using the sliding window method, and each training sample includes H airspace traffic; each training sample is used as the input of the traffic prediction model, and the airspace traffic after the predicted sliding step size L is used as the output, and the actual airspace traffic after each training sample is used as the prediction target. The prediction result is evaluated for model accuracy using the mean absolute percentage error MAPE. When the calculated MAPE is less than the preset MAPE, the traffic prediction model training corresponding to the airspace traffic is completed; wherein, the calculation formula of the mean absolute percentage error MAPE is: ; In the formula, For the The predicted target corresponding to the predicted airspace traffic is For the The predicted airspace traffic, The number of predicted airspace traffic flows; a traffic prediction model is generated based on the airspace traffic flows to predict the airspace traffic flows at future times.

[0090] For example, the traffic training set A includes 10 airspace traffics. , For the Airspace traffic, , define the length of the sliding window as 3, the sliding step L as 1, use the sliding window to construct 7 training samples, each training sample contains 3 consecutive airspace flows, and the next airspace flow of the 3 consecutive airspace flows is used as the prediction target; for example: training sample , training samples The corresponding prediction target is ; Training samples , training samples The corresponding prediction target is ; And so on, it is used to train the traffic prediction model.

[0091] The compensation payment module calculates the compensation amount based on the flight delay duration and determines whether to pay meals.

[0092] The methods for calculating the compensation amount and making compensation payments include:

[0093] The compensation standard is preset by the technicians in this field according to the actual situation. The compensation standard includes the compensation amount corresponding to different flight delay durations. For example, if the flight delay duration is less than 60 minutes, no compensation is made; if the flight delay duration is between 60 minutes and 460 minutes, the compensation amount is 200 yuan; if the flight delay duration exceeds 460 minutes, the compensation amount is 400 yuan; according to the compensation standard and the flight delay duration, the corresponding compensation amount is obtained;

[0094] Obtain the passenger payment information corresponding to the target flight. The passenger payment information includes the payment type and payment account. The passenger payment information is obtained through the ticket purchasing system of the airline corresponding to the target flight. The payment type is the payment method selected by the passenger when purchasing the ticket, such as credit card payment, savings card, WeChat payment, Alipay payment, etc. The payment account is the account information associated with the payment type when the passenger purchases the ticket, such as mobile phone number, bank card number, etc.; based on the passenger payment information corresponding to the target flight, the compensation amount is paid to all passengers corresponding to the target flight.

[0095] Methods for determining whether to distribute meals include:

[0096] Preset the issuance threshold, which is preset by technical personnel in this field according to actual conditions; compare the flight delay duration with the issuance threshold; if the flight delay duration is greater than or equal to the issuance threshold, meals are issued, because a long waiting time will make passengers anxious and uneasy, and feel hungry, and providing meals can help soothe passengers and reduce hunger; if the flight delay duration is less than the issuance threshold, add the current time to the flight delay duration to obtain the estimated departure time; preset meal time, which includes breakfast time , lunch time and dinner time ,in, ; The meal time is pre-set by the technicians in charge according to the actual situation; compare the estimated departure time with the meal time; if , then meals will be distributed, indicating that passengers will experience meal time while waiting for the flight, which will cause passengers to feel hungry and uncomfortable, so meals need to be provided to ensure that passengers can get the necessary dietary supplements during the waiting process. is the estimated time of departure, means or; if , no meals will be distributed, which means that the passengers have experienced a short waiting time and have not passed the meal time. Providing meals will lead to a waste of resources.

[0097] If meals are to be distributed, the mobile phone number of each target passenger is obtained and marked as a push number, and the target passenger is the passenger corresponding to the target flight; the mobile phone number of each target passenger is obtained through the ticket purchasing system of the airline corresponding to the target flight; based on the push number of each target passenger, an electronic meal coupon is pushed to each target passenger; the method of pushing the electronic meal coupon may be WeChat applet push, SMS push, etc.; the electronic meal coupon is a digital meal voucher, and passengers can use the electronic meal coupon to choose their own meals in the restaurant in the terminal, thereby solving the problems of long waiting time, single taste, and limited distribution time in traditional flight delay meal services, and thus meeting the diverse dining needs of passengers.

[0098] The accommodation arrangement module determines whether to generate arrangement information based on the flight delay duration.

[0099] The method of determining whether to generate arrangement information includes:

[0100] A scheduling threshold is preset, and the scheduling threshold is preset by technical personnel in this field according to actual conditions; the estimated departure time is compared with the scheduling threshold; if the estimated departure time is greater than or equal to the scheduling threshold, scheduling information is generated, indicating that the estimated departure time of the flight is late and the passengers cannot reach the destination on the same day, so the airline needs to arrange accommodation and transportation for the target passengers; if the estimated departure time is less than the scheduling threshold, no scheduling information is generated.

[0101] Arrangement information includes transportation arrangement information and accommodation arrangement information;

[0102] Methods for generating traffic arrangement information include:

[0103] Obtain the passenger information data of each target passenger, including age, cabin type, check-in counter and check-in time; cabin type is the class corresponding to the ticket purchased by the passenger, check-in counter is the service window for the passenger to check in, check-in time is the time when the passenger checks in before the flight takes off, and the unit of check-in time is consistent with the unit of current time; cabin types include first class, business class and economy class; the passenger information data of each target passenger is obtained through the check-in system of the airline corresponding to the target flight; according to the check-in counter and check-in time of each target passenger, all target passengers are divided into travel groups, among which if only one passenger travels, they are also considered as a travel group;

[0104] Subtract the check-in time of each target passenger from the current time to obtain the waiting time of each target passenger; calculate the priority level of each travel group based on the age, cabin type and waiting time of each target passenger; divide all travel groups into group sets based on the priority level of each travel group; set a numerical label for each group set in ascending order and mark it as a group label; obtain a single round trip time, which is the time it takes for the shuttle vehicle to pick up the passengers to the accommodation and then return to the terminal; multiply each single round trip time by the group label corresponding to each group set, and add the result to the current time to obtain the travel time corresponding to each group set; convert the travel time corresponding to each group set into a standard time format (24-hour system) and use it as transportation arrangement information, and push the transportation arrangement information corresponding to each group set to the corresponding target passengers. The method of pushing transportation arrangement information is the same as the method of pushing electronic meal coupons.

[0105] Methods for dividing all target passengers into travel groups include:

[0106] The target passengers corresponding to each check-in counter are taken as a check-in set, and the check-in set corresponds to the check-in counter one by one; each target passenger in each check-in set is sorted from large to small according to the corresponding check-in time, and a check-in sorting table corresponding to each check-in set is generated; the difference between each two adjacent check-in times in each check-in sorting table is calculated in turn as the check-in time difference; a time difference threshold is preset, and the time difference threshold is preset by technical personnel in this field according to actual conditions; each check-in time difference is compared with the time difference threshold respectively; if the check-in time difference is less than or equal to the time difference threshold, the two target passengers corresponding to the corresponding check-in time difference are taken as a travel group; if the check-in time difference is greater than the time difference threshold, the two target passengers corresponding to the corresponding check-in time difference are each taken as a travel group; all travel groups are analyzed, and if there are the same target passengers in two travel groups, the travel groups with the same target passengers are merged until there are no same target passengers in all travel groups.

[0107] The methods for calculating the priority corresponding to each travel group include:

[0108] Different cabin types are assigned different digital numbers and marked as cabin numbers. The cabin numbers are pre-set by technicians in this field according to the prices corresponding to different cabin types. Cabin types with high prices have large cabin numbers, while cabin types with low prices have small cabin numbers. Therefore, the cabin number corresponding to first class is larger than the cabin number corresponding to business class, and larger than the cabin number corresponding to economy class. Preset age threshold and , , ; Compare the age of each target passenger with the age threshold and For comparison; if , then the age coefficient assigned to the corresponding target passenger is 0, is the age of the target passenger; if , then the age coefficient assigned to the corresponding target passenger is ;like , then the age coefficient assigned to the corresponding target passenger is ;

[0109] The expression of the priority corresponding to the target passenger is: ; In the formula, For priority, is the age coefficient, is the cabin number, The waiting time is , , are all preset weight coefficients; the specific values ​​of the weight coefficients can be set according to actual conditions, and the weight coefficients reflect the influence of the age coefficient, cabin number and waiting time on the priority. Technical personnel in this field can preset corresponding weight coefficients according to the actual influence of the age coefficient, cabin number and waiting time on the priority, so as to accurately evaluate the priority of different target passengers.

[0110] It should be understood that the age coefficient, cabin number and waiting time are all factors affecting priority; among them, the larger the age coefficient, the younger the child or the older the elderly, who usually have weaker physical and mental adaptability to travel and need more care and attention, so they need to be given priority when arranging accommodation and transportation; the larger the cabin number, the more the passenger pays for the ticket and should enjoy higher service standards, so they should enjoy priority; the longer the waiting time, the lower the passenger's patience and comfort level tend to be, and long waits can cause passengers to feel uneasy or tired, so giving priority to these passengers can effectively improve the passenger experience.

[0111] Count the number of target passengers in each travel group and mark it as the number of passengers; add up the priorities of all target passengers in each travel group in turn, and then divide it by the corresponding number of passengers to obtain the priority corresponding to each travel group.

[0112] Methods for dividing all travel groups into groups include:

[0113] Get the number of shuttle vehicles and the vehicle passenger capacity. Shuttle vehicles are transportation tools (such as airport buses, vans, taxis, etc.) that transport passengers to accommodation places (such as hotels, inns, homestays, etc.), and the vehicle passenger capacity is the maximum number of passengers that a single shuttle vehicle can transport. The number of shuttle vehicles and the vehicle passenger capacity are manually reported by the relevant staff of the airline. Sort each travel group from large to small according to the corresponding priority to generate a group sorting table. Replace each travel group in the group sorting table with the corresponding number of passengers. Add all the passengers in front of the corresponding number of passengers to each passenger number in the group sorting table to obtain the total number of passengers corresponding to each passenger number, and replace each passenger number in the group sorting table with the corresponding total number of passengers. Multiply the number of shuttle vehicles by the vehicle passenger capacity to obtain the maximum single passenger capacity. Mark the total number of passengers with the largest value in the group sorting table as the maximum number of passengers. Divide the maximum number of passengers by the maximum single passenger capacity and round up to obtain the number of connections. Multiply the maximum single passenger capacity by the increasing connection coefficient to obtain The overall maximum passenger capacity; where the connection coefficient is an integer and the range of the connection coefficient is , n is the number of connections; The total maximum passenger capacity is added to the group sorting table, and the travel group corresponding to the total number of passengers between every two total maximum passenger capacities is taken as a group set to obtain n group sets; the range of group labels is .

[0114] Methods for generating accommodation information include:

[0115] Count the number of people in each travel group who are younger than the age threshold The target number of passengers is obtained and marked as the exclusion number; the number of passengers in each travel group is subtracted from the corresponding exclusion number to obtain the number of accommodations corresponding to each travel group; the number of accommodations for each travel group is divided by 2 to obtain the number of rooms corresponding to each travel group; each travel group in the group sorting table is replaced with the corresponding number of rooms, and the corresponding rooms are arranged for each travel group in positive order, and the room number corresponding to each travel group is obtained; the room number corresponding to each travel group is used as the accommodation arrangement information, and the accommodation arrangement information corresponding to each travel group is pushed to the corresponding target passengers. The method of pushing the accommodation arrangement information is the same as the method of pushing the electronic meal voucher.

[0116] By collecting and analyzing aviation operation data, this embodiment can accurately obtain route traffic and use deep learning technology to predict the probability of flight delays in real time, so as to provide timely flight delay risk warnings; according to the generated warning instructions, accurate compensation, meal distribution, transportation arrangements and accommodation arrangements are provided to effectively respond to flight delays and solve the problem that traditional manual management and single emergency response methods cannot accurately meet passenger needs; big data and artificial intelligence technologies are fully utilized to realize intelligent prediction of flight delay risks and refined services, which not only improves the operating efficiency and service efficiency of airlines, but also greatly improves passengers' travel experience and satisfaction, providing strong support for the intelligent management of the aviation industry.

[0117] Example 2

[0118] The present application also provides an electronic device. The electronic device may include one or more processors and one or more memories. The memories store computer-readable codes, and when the computer-readable codes are executed by one or more processors, the flight delay risk warning and response system based on big data analysis as described above may be executed.

[0119] The method or system according to the implementation mode of the present application can also be implemented with the help of the architecture of the electronic device shown in the present application. The electronic device may include a bus, one or more CPUs, ROM, RAM, a communication port connected to the network, input / output, a hard disk, etc. A storage device in the electronic device, such as a ROM or a hard disk, can store the flight delay risk warning and response system based on big data analysis provided by the present application. Furthermore, the electronic device may also include a user interface. Of course, the architecture shown in the present application is only exemplary. When implementing different devices, one or more components in the electronic device shown in the present application may be omitted according to actual needs.

[0120] Example 3

[0121] One embodiment of the present application discloses a computer-readable storage medium. Computer-readable instructions are stored on the computer-readable storage medium. When the computer-readable instructions are executed by the processor, the flight delay risk warning and response system based on big data analysis according to the embodiment of the present application described with reference to the above figures can be executed. The storage medium includes, but is not limited to, for example, volatile memory and / or non-volatile memory. Volatile memory may include, for example, random access memory (RAM) and cache memory (cache), etc. Non-volatile memory may include, for example, read-only memory (ROM), hard disk, flash memory, etc.

[0122] In addition, according to the embodiments of the present application, the process described above with reference to the flowchart can be implemented as a computer software program. For example, the present application provides a non-transitory machine-readable storage medium, which stores machine-readable instructions, and the machine-readable instructions can be executed by a processor to execute instructions corresponding to the method steps provided in the present application, such as: a flight delay risk warning and response system based on big data analysis. When the computer program is executed by a central processing unit (CPU), the above functions defined in the method of the present application are executed.

[0123] The above is only a specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art can easily think of changes or substitutions within the technical scope disclosed by the present invention, which should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention should be based on the protection scope of the claims.

[0124] Finally: The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. Flight delay risk warning and response system based on big data analysis, characterized by: include: Data collection module, used to collect aviation operation data; Data analysis module, used to analyze aviation operation data and obtain route traffic; Delay prediction module, which is used to analyze aviation operation data and route traffic to predict the probability of flight delays; The delay warning module determines whether to generate a warning instruction based on the probability of flight delay. If a warning instruction is generated, the flight delay duration is predicted; The compensation payment module calculates the compensation amount based on the flight delay duration and determines whether to pay meals. The accommodation arrangement module determines whether to generate arrangement information based on the flight delay duration.

2. The flight delay risk warning and response system based on big data analysis according to claim 1 is characterized in that: The aviation operation data includes flight data and weather data; Flight data collection methods include: Collect target flight data, which is the flight information of the target flight; the target flight is the flight for which flight delay prediction is performed, and the flight information includes take-off time, landing time, flight route and flight distance, the take-off time is the time when the flight is scheduled to take off from the departure airport, the landing time is the time when the flight is scheduled to land at the arrival airport, the flight route is the flight path between the departure airport and the arrival airport of the flight, and the flight distance is the flight distance corresponding to the flight route; obtain the current time, and mark the take-off time in the target flight data as the target take-off time; according to the current time and the target take-off time, obtain related flights, which are all flights whose take-off times are between the current time and the target take-off time; according to the related flights, obtain related flight data, which is the flight information of the related flights; Mark the flight routes in the target flight data as target routes, and mark the flight routes in the related flight data as related routes; obtain a world map, draw the target route and the related routes on the world map, and determine whether there is a route in the related routes that intersects with the target route; mark the routes in the related routes that intersect with the target route as related routes, and do not mark the routes in the related routes that do not intersect with the target route; mark the flight corresponding to the related route as a related flight, mark the flight information of the related flight as related flight data, and use the related flight data and the target flight data as flight data; The weather data is the weather information corresponding to the departure airport at the target take-off time, including wind speed, precipitation, visibility and temperature.

3. The flight delay risk warning and response system based on big data analysis according to claim 2 is characterized in that: The method for obtaining route traffic includes: Mark the take-off time in the associated flight data as the associated take-off time, and mark the landing time as the associated landing time; convert the current time, target take-off time, associated take-off time, and associated landing time into a unified unit; subtract the corresponding associated take-off time from the associated landing time of each associated flight to obtain the estimated flight time of each associated flight; subtract the associated take-off time of each associated flight from the target take-off time to obtain the elapsed flight time of each associated flight; divide the elapsed flight time of each associated flight by the corresponding estimated flight time, and then multiply by the corresponding flight distance to obtain the elapsed flight distance of each associated flight; Get the departure airport and landing airport of each associated flight. According to the world map, get the longitude and latitude corresponding to the departure airport and landing airport of each associated flight. The longitude and latitude include longitude and latitude. Mark the longitude and latitude corresponding to the departure airport as the starting longitude and latitude, and mark the longitude and latitude corresponding to the landing airport as the ending longitude and latitude. Subtract the corresponding starting longitude from the ending longitude of each associated flight to obtain the longitude difference. Subtract the corresponding starting latitude from the ending latitude of each associated flight to obtain the latitude difference. Calculate the current longitude and latitude corresponding to each associated flight according to the longitude difference, latitude difference, flown distance and flight distance corresponding to each associated flight. The expression of the current longitude is: ; In the formula, is the current longitude, is the starting point longitude, is the distance flown, is the flight distance, is the longitude difference; the expression for the current latitude is: ; In the formula, is the current latitude, is the starting point latitude, is the latitude difference; Divide the world map into airspace according to the preset division scheme to obtain airspace, is an integer greater than 1; according to the current longitude and latitude of each associated flight, mark the position of each associated flight on the world map; mark the airspace passed by the target route on the world map as the target airspace, count the number of associated flights in each target airspace, and mark it as airspace traffic; take the airspace traffic corresponding to all target airspaces as route traffic.

4. The flight delay risk warning and response system based on big data analysis according to claim 3 is characterized in that: The method for predicting the probability of flight delay comprises: Use weather data and route traffic as analysis data, input the analysis data into the trained delay prediction model, and predict the corresponding flight delay probability; The method for determining whether to generate an early warning instruction includes: Set a warning threshold and compare the flight delay probability with the warning threshold; if the flight delay probability is greater than or equal to the warning threshold, generate a warning instruction; if the flight delay probability is less than the warning threshold, do not generate a warning instruction; The step of predicting flight delay duration includes: Step 1: Obtain historical traffic, which is the route traffic obtained at historical time; take the airspace traffic corresponding to the same target airspace in the historical traffic as a set of traffic sets, and the traffic set corresponds to the target airspace one by one; Step 2: Input each traffic set into the trained traffic prediction model to predict the airspace traffic at the future time T, where T is an integer greater than 1; Step 3: Take all predicted airspace traffic as predicted traffic, obtain weather data at future times, predict the corresponding flight delay probability based on the weather data at future times and predicted traffic, and determine whether to generate an early warning instruction; Step 4: If a warning instruction is still generated, replace the earliest airspace traffic in each group of traffic sets with the corresponding predicted airspace traffic, and return to step 2; if no warning instruction is generated, go to step 5; Step 5: Get the number of warning instructions generated in step 3 and mark it as the delay number. Multiply the delay number by T to get the flight delay duration. The unit of the flight delay duration is consistent with the unit of the target take-off time.

5. The flight delay risk warning and response system based on big data analysis according to claim 4 is characterized in that: In step 2, the traffic prediction model is an RNN neural network model, and the training process of the traffic prediction model includes: Pre-collect the same airspace continuously airspace traffic and build a traffic training set. , is the number of airspace flows in a set of flow sets; based on the flow training set, a flow prediction model for predicting airspace flows at future moments is trained; Preset the sliding step size L and the sliding window length H; use the sliding window method to convert the airspace traffic in the traffic training set into multiple training samples, each training sample includes H airspace traffic; use each training sample as the input of the traffic prediction model, predict the airspace traffic after the sliding step size L as the output, and use the actual L airspace traffic corresponding to each training sample as the prediction target. Use the mean absolute percentage error MAPE to evaluate the model accuracy of the prediction results. When the calculated MAPE is less than the preset MAPE, the traffic prediction model training corresponding to the airspace traffic is completed; generate a traffic prediction model that predicts the airspace traffic at future times based on the airspace traffic.

6. The flight delay risk warning and response system based on big data analysis according to claim 5 is characterized in that: The method for calculating the compensation amount and making compensation payment includes: Preset compensation standards, including compensation amounts corresponding to different flight delay durations; obtain corresponding compensation amounts according to the compensation standards and flight delay durations; obtain passenger payment information corresponding to the target flight, including payment type and payment account number, and pay the compensation amount to all passengers corresponding to the target flight according to the passenger payment information corresponding to the target flight; The method for determining whether to distribute meals includes: Preset the delivery threshold and compare the flight delay duration with the delivery threshold; if the flight delay duration is greater than or equal to the delivery threshold, the meal will be delivered; if the flight delay duration is less than the delivery threshold, the current time will be added to the flight delay duration to obtain the estimated departure time; preset meal time, meal time includes breakfast time , lunch time and dinner time ,in, ; Compare the estimated departure time with the meal time; if , meals will be distributed, including is the estimated time of departure, means or; if , no meals will be distributed; If meals are distributed, the mobile phone number of each target passenger is obtained and marked as a push number. The target passengers are passengers corresponding to the target flights. Electronic meal coupons are pushed to each target passenger based on the push number of each target passenger.

7. The flight delay risk warning and response system based on big data analysis according to claim 6 is characterized in that: The method for determining whether to generate arrangement information includes: A scheduling threshold is preset, and the estimated take-off time is compared with the scheduling threshold; if the estimated take-off time is greater than or equal to the scheduling threshold, scheduling information is generated; if the estimated take-off time is less than the scheduling threshold, scheduling information is not generated; The arrangement information includes transportation arrangement information and accommodation arrangement information; Methods for generating traffic arrangement information include: Obtain the passenger information data of each target passenger, which includes age, cabin type, check-in counter and check-in time, and the unit of check-in time is consistent with the unit of current time; divide all target passengers into travel groups according to the check-in counter and check-in time of each target passenger, where if only one passenger travels, it is also considered as a travel group; subtract the check-in time of each target passenger from the current time to obtain the waiting time of each target passenger; calculate the priority corresponding to each travel group according to the age, cabin type and waiting time of each target passenger; divide all travel groups into group sets according to the priority of each travel group; set digital labels for each group set in increasing order and mark them as group labels; obtain the single round-trip time, which is the time it takes for the shuttle vehicle to pick up the passenger to the accommodation and then return to the terminal; multiply each single round-trip time by the group label corresponding to each group set, and add the current time to obtain the boarding time corresponding to each group set; convert the boarding time corresponding to each group set into a standard time format, and use it as traffic arrangement information, and push the traffic arrangement information corresponding to each group set to the corresponding target passenger.

8. The flight delay risk warning and response system based on big data analysis according to claim 7 is characterized in that: The method for dividing all target passengers into travel groups comprises: The target passengers corresponding to each check-in counter are regarded as a group of check-in sets, and the check-in sets correspond to the check-in counters one by one; each target passenger in each check-in set is sorted from large to small according to the corresponding check-in time, and a check-in sorting table corresponding to each check-in set is generated; the difference between each two adjacent check-in times in each check-in sorting table is calculated in turn as the check-in time difference; a time difference threshold is preset, and each check-in time difference is compared with the time difference threshold respectively; if the check-in time difference is less than or equal to the time difference threshold, the two target passengers corresponding to the corresponding check-in time difference are regarded as a travel group; if the check-in time difference is greater than the time difference threshold, the two target passengers corresponding to the corresponding check-in time difference are regarded as a travel group respectively; all travel groups are analyzed, and if there are the same target passengers in two travel groups, the travel groups with the same target passengers are merged until there are no same target passengers in all travel groups; The method for calculating the priority corresponding to each travel group includes: Set different digital numbers for different cabin types and mark them as cabin numbers; preset age thresholds and , , ; Compare the age of each target passenger with the age threshold and For comparison; if , then the age coefficient assigned to the corresponding target passenger is 0, is the age of the target passenger; if , then the age coefficient assigned to the corresponding target passenger is ;like , then the age coefficient assigned to the corresponding target passenger is ; The expression of the priority corresponding to the target passenger is: ; In the formula, For priority, is the age coefficient, is the cabin number, The waiting time is , , are all preset weight coefficients; the number of target passengers in each travel group is counted and marked as the number of passengers; the priorities of all target passengers in each travel group are added up in turn, and then divided by the corresponding number of passengers to obtain the priority corresponding to each travel group.

9. The flight delay risk warning and response system based on big data analysis according to claim 8 is characterized in that: The method for dividing all travel groups into group sets includes: Get the number of shuttle vehicles and the vehicle passenger capacity, where the vehicle passenger capacity is the maximum number of passengers that a single shuttle vehicle can transport; sort each travel group from large to small according to the corresponding priority to generate a group sorting table; replace each travel group in the group sorting table with the corresponding number of passengers; add each passenger number in the group sorting table to all the passengers in front of the corresponding passenger number to obtain the total number of passengers corresponding to each passenger number, and replace each passenger number in the group sorting table with the corresponding total number of passengers; multiply the number of shuttle vehicles by the vehicle passenger capacity to obtain the maximum single passenger capacity; mark the total number of passengers with the largest value in the group sorting table as the maximum number of passengers, divide the maximum number of passengers by the maximum single passenger capacity, and round up to obtain the number of connections; multiply the maximum single passenger capacity by the increasing connection coefficient to obtain The overall maximum passenger capacity; where the connection coefficient is an integer and the range of the connection coefficient is , n is the number of connections; The total maximum passenger capacity is added to the group sorting table, and the travel group corresponding to the total number of passengers between every two total maximum passenger capacities is taken as a group set to obtain n group sets; the range of group labels is .

10. The flight delay risk warning and response system based on big data analysis according to claim 9, characterized in that: Methods for generating accommodation information include: Count the number of people in each travel group who are younger than the age threshold The target number of passengers is obtained and marked as the exclusion number; the number of passengers in each travel group is subtracted from the corresponding exclusion number to obtain the number of accommodations corresponding to each travel group; the number of accommodations for each travel group is divided by 2 to obtain the number of rooms corresponding to each travel group; each travel group in the group sorting table is replaced with the corresponding number of rooms, and the corresponding rooms are arranged for each travel group in positive order, and the room number corresponding to each travel group is obtained; the room number corresponding to each travel group is used as the accommodation arrangement information, and the accommodation arrangement information corresponding to each travel group is pushed to the corresponding target passengers.

Images